Regression Test Selection for Android Applications
Do, Quan Chau Dong
As the mobile platform pervades human life, much research in recent years has focused on improving the reliability of mobile applications on this platform, for example by applying automatic testing. To date, however, researchers have primarily considered testing of the single version of mobile applications. It has been shown that testing of mobile applications can be expensive; thus simply re-executing all tests on the modified application version remains challenging. Regression testing---a process of validating modified software to ensure that the changes are correct and do not adversely affect other features of the software---has been extensively studied for desktop application, and many efficient and effective approaches have been proposed; however, these approaches cannot be directly applied to mobile applications. Since regression testing on mobile applications is an expensive process, an effective and well-studied regression test selection can potentially reduce this expense. In this study, we propose test selection for mobile applications, especially on the Android Application Platform. Our approach leverages the combination of static impact analysis with code coverage that is dynamically generated at run-time, and identify a subset of tests to check the behaviors of the modified version that can potentially be different from the original version. We implement our approach for Google Android applications, and demonstrate its effectiveness using an empirical study.
Regression, Testing, Android, Test selection
Do, Q. C. D. (2015). <i>Regression test selection for android applications</i> (Unpublished thesis). Texas State University, San Marcos, Texas.